In recent decades, water scarcity has worsened over large areas of India as a result of demand for water outstripping sustainable supply. Underlying causes include: population growth, a growing economy, land use change, urbanisation and intensification of rainfed and irrigated agriculture. The risk also exists that water scarcity will be exacerbated by climate change. The most serious impact is likely to be an increase in the frequency and severity of extreme events, such as floods and droughts, which already affect millions of people across India with alarming regularity.
Aquifers have become the primary source of water for both agricultural and domestic use in many rural areas of India, both due to increasing demand and the limited and seasonal nature of surface water availability. This increase in groundwater use has also been driven by improvements in well technology, decreases in construction costs, and government subsidies for electricity and diesel fuel. One result has been a large expansion of the area of land under irrigation, which, along with other agricultural advances, such as increased fertiliser and pesticide use and the adoption of new crop varieties, has improved agricultural productivity and profitability.
However, subsidised and unregulated use of groundwater has led to the steady decline of the water table in many regions and the widespread failure of wells1. To try and maintain access to water, farmers have had to increase both the number and depth of borewells they are constructing. Poorer farmers, often from lower social/caste groups, are generally less able to do so and have therefore seen their access to water curtailed. Well failure and a subsequent inability to repay construction loans has been cited as a primary cause of the high suicide rates of farmers in rural India and a factor in driving migration from rural to urban areas.
Falling groundwater levels have also increased pressure on domestic water supplies as many traditional drinking water sources can no longer be relied on. A large number of rural households do not receive the minimum 40 litres per capita per day of clean water recommended by the Government of India and the World Health Organisation. As a result, poorer households in particular, continue to face significant social, economic and health related problems2.
The WASHCost project, funded by the Bill and Melinda Gates Foundation, worked in India, Burkina Faso, Ghana and Mozambique during 2008-2012 on the theoretical and actual costs of sustainable, equitable and efficient Water, Sanitation and Hygiene (WASH) service provision . In India, it became clear that WASH service levels were influenced significantly by factors that included: geographical location, agricultural water use, activities aimed at intensifying irrigated and rainfed agriculture, social status, and various political economy factors. These preliminary findings prompted a detailed study of patterns of water accessibility, use and competition in a number of revenue villages across Andhra Pradesh.
The visualisations presented in this case study use data from a revenue village located a few kilometres west of the town of Dhone in Kurnool district. It has an area of 25.4 km2 and consists of 4 individual villages that in 2012 had a total of 1087 households (i.e. a total population of around 5,500). Single and double crop agriculture accounts for approximately 60% and 8% respectively of the revenue village’s total land area. However, these figures vary significantly depending on rainfall in any given year. Other important land uses included horticultural plantations, primarily mango, which accounts for 11% of the area, and scrub forest, which accounts for 16%.
The region is semi-arid, with an average annual rainfall of 618 mm. The majority of this rainfall is contributed by the southwest monsoon, between June and September, and the northeast monsoon, between October and December. The varying strengths of the monsoons leads to large variations in annual and monthly rainfall totals. The rainfall visualisations below use 26 years of monthly and 11 years of daily rainfall data collected at the meteorological station in Dhone town, a few kilometres outside the revenue village boundary.
As can be seen in the daily rainfall visualisation, a relatively small number of days often account for a large proportion of monthly and annual rainfall. For example, 89.8 and 246.8 mm of rain fell respectively on the 23rd and 24th June 2007, which, in total, accounted for 66% of the monthly and 23% of the annual rainfall totals. More generally, of the 7450 mm of rain that occurred in the 11-year period between June 1997 and May 2008, 27% fell on days with 50 mm of rainfall or more. However, this figure varied significantly between wet and dry years. In years with below average rainfall, 11% fell on days with more than 50mm of rainfall, whereas for years with above average rainfall the figure was 33%.
The geology of the revenue village is crystalline basement, which generally consists of fractured bedrock overlain by weathered material. Groundwater can be found both in this weathered material and in cracks and fissures in the bedrock. While borewells drilled into the bedrock can provide a reliable supply of water, they also frequently fail when dug in areas where the bedrock is relatively unbroken3. As bedrock characteristics can vary significantly over short distances, successfully siting a borewell is often a challenge.
Managing water resources in semi-arid rural India is complicated by a large number of biophysical and societal factors and the fact that different stakeholders have a range of different agricultural, social, economic, political, and environmental goals. A further complication is that for many catchments the quantity, location, ownership and status of water infrastructure, including wells and Rain Water Harvesting (RWH) structures, is unknown or poorly documented. One solution is to use GPS technology and participatory approaches to map the location of infrastructure, a task increasingly carried out using smartphone apps, such as Akvo Flow and mWater.
For this study, standard handheld GPS devices were used due to their relative robustness and simplicity in comparison to smartphone options. Each well and RWH structure located in the revenue village was mapped, and additional information was collected including type, cost, year of construction, ownership and status. Most of the survey work was carried out by four students from the revenue village who were contracted by the study and supervised by a retired hydro-geologist from the state groundwater department.
The visualisation below shows how the number and characteristics of wells located in the revenue village have changed over time.
The most dramatic trend shown by the visualisation is the rapid increase in the number of wells, particularly since the turn of the century. In 1980 there were 86 wells in the revenue village at an average density of 3.4 per km2. By 2012, the number had increased to 573 at a density of 22.6 per km2. Between 1990 and 2000, the average number of wells constructed each year was 8.5, this rose to 22.7 per year between 2000 and 2006, to 39 per year between 2006 and 2012. This acceleration in well construction was driven by the profitability of irrigated agriculture relative to rainfed agriculture and the failure of older and shallower wells due to falling groundwater levels.
The visualisation also shows that the increase in well numbers was coupled with an increase in their depth. The average depth of wells constructed prior to 1980 was 8.7 metres. This increased to 37 metres between 1980 and 2000, to 50.7 metres between 2000 and 2006, and finally to 64 metres between 2006 and 2012. Also notable is a shift in the type of well being constructed. Of all the wells constructed prior to the year 2000 only 32% were borewells. Between 2000 and 2012 this proportion rose to 82%.
It is clear from the visualisation that failure has not just been a problem for shallow open wells. Of the 327 wells mapped in the revenue village that were more than 50 metres deep, 38% had failed to some extent, including 26 wells over 100 metres deep. On the other hand, of the 198 wells that were mapped with depths of less than 10 metres, 20% had never failed and only 35% had failed completely, which indicates that, in addition to depth, location and other factors are important in determining well reliability.
In general, the overexploitation of groundwater has impacted negatively on household water service levels in the revenue village. A detailed survey indicated that 58% of households received less than 40 litres of water per capita per day during either or both of the summer and non-summer seasons. During the non-summer season households spent an average of 60 minutes per day collecting water. This rose to 95 minutes during the summer as water sources became less reliable or failed completely.
The impacts of water scarcity have not been felt evenly by all groups and individuals. Households with lower per capita water use and that spent more time collecting water were generally those that had lower income and land ownership than average and that were members of lower social/caste groups. Furthermore, within many households the task of collecting water falls disproportionately on women and children, impacting their health and education.
The primary solution to groundwater depletion in India has been the construction of RWH structures that capture surface runoff and increase localised groundwater recharge. Other stated benefits of the structures include the attenuation of downstream flood peaks and reductions in the sedimentation of downstream reservoirs.
Many RWH structures in rural areas of India have been funded and built by the government’s watershed development programmes. Since the 1980s, these programmes have invested billions of dollars in structures and other measures, however quantitative evidence of widespread benefits, in terms of improving water security, is relatively scarce4. This is partly because measuring the effectiveness of RWH structures at local scales is a challenging and time-consuming task5. Nevertheless, RWH structures are seen as an essential tool for improving groundwater sustainability and if well designed and sited can have significant local benefits6.
In many areas, local NGOs and community groups have been influential in promoting RWH and other conservation measures, such as reforestation. One of the best known is the Tarun Bharat Singh organisation, which has helped to rejuvenate the Arvari catchment in Rajasthan via the promotion of RWH and the formation of a river parliament, through which 72 villages can communally manage the catchment’s water resources. Another is the Timbuktu collective, which has undertaken similar work in Anantapur district in southern Andhra Pradesh.
The visualisation below shows details of the RWH structures located in the revenue village.
In total 55 check dams, 42 infiltration ponds, and 6 bunds have been constructed in the revenue village at a total cost of around $55,000. The majority have been built during the last two decades, mostly in years when government projects have prioritised and focused on the area. The total capacity of the RWH structures that have not been destroyed or fallen into disrepair is roughly 114,000 m3, which is equivalent to approximately 4 mm of rainfall spread evenly across the catchment. However, as discussed earlier, a large proportion of rainfall occurs over a small number of days, which limits how much water can be captured by the structures. During extreme rainfall events, structures quickly fill up and overflow and, in some cases, can be damaged or destroyed.
Overall, the success of RWH has been mixed, with only 35% of the structures in good condition. One issue is siltation, which has significantly reduced the capacity of some of the older structures. This suggests that RWH needs to be better supported by other measures, such as the reforestation of the surrounding hills, to reduce soil erosion. On a positive note, the RWH structure with the largest capacity plays an important role in maintaining the reliability of the 3 borewells that supply the majority of the drinking water used by village of Ungaranigundla.
The effectiveness of RWH structures depends heavily on local factors, particularly the efficiency of groundwater recharge, the storage capacity of the underlying aquifers, and the dynamic interactions between surface water and groundwater7. However, these factors are often ignored in the planning of watershed interventions, as is the fact that by potentially reducing downstream flows, intensive RWH can increase inequality in catchments in terms of access to water. This can be seen in the revenue village by the fact that some RWH structures have been purposely damaged or destroyed to allow more water to flow downstream.
The impact of RWH on downstream flows tends to be greatest in years with low rainfall or during prolonged periods of drought, which are normally times of heightened demand for water downstream8. With conflict between states over the sharing of water resources already increasing in the region, the impact of intensified RWH, as well as climate change, could further inflame tensions.
More equitable sharing of water resources, both within the revenue village and downstream, requires increased focus on demand management, in particular, to reduce the overexploitation of groundwater. While RWH has an important role to play, it needs to be carefully planned so as to avoid significant downstream externalities. In this regard, leveraging new technology, such as GPS-enabled smartphones, to map and collect information about structures, can help improve decision-making, especially as advances in web technologies allow for better visualisation and communication of such data. Inevitably, trade-offs and compromises will be required, but there are an increasing number of examples, such as in the Arvari catchment9, that demonstration how community management can help to reduce the impacts of water scarcity on rural communities in India.